One of my customers asked me if would take over maintenance and development of their Progress accounting system. They have all the source code, but no system docs. Any thoughts?

What's "Progress accounting system"?
I would do some sort of review of the source code to decide how much I wanted the job, and then quote for the work according to sliding scale of how much of a PITA it looks like. No system docs is fine so long as part of the "maintenance and development" includes a long time writing them.

pootle flump ur codings are working excelent.

Charge them for evaluating the source, then make a decision based on what you found.

Pfffft - you just cribbed me.

pootle flump ur codings are working excelent.

A lot depends on issues that you could never cover in a post.

From a technical side, I'd buy a copy of SQL Spec if you don't already have one. It makes nifty documentation of the Progress database, and this is where the surprises usually lurk. I take this back, SQL Spec does NOT document Progress databases... I thought that I had done that, but I must be corn-fused.

From an accounting side, how picky is your client and how comfortable are you with GAAP and specifically with Canadian legislation in that area? Unless you are really, REALLY good I'd be sure to write the contract or statement of work so that you're only responsible for implementing what the client requests. There is WAY too much liability for this to be profitable any other way that I can imagine.

If you want, I'll be happy to pull this into a new thread. Otherwise we'll both be lambasted for talking shop in the corral!

-PatP

ready teddy's sig. one of my specialties is a field I have coined software archealogy.

Pat, can you set up a new thread?
Thanks, Jack

Likewise on this side of the fence. My few years of spelunking around legacy source code in an attempt to extract meaningful architecture cues has taught me that source access can be damn near useless when stacked next to the cost of a rewrite or off the shelf purchase. Or, it could be completely self-documenting and read like a tv guide. You just never know...

Done!

-PatP

I wouldn't go near it

Oh come on! No guts, no glory!

While Canada is no where near as litagious as the United States, there is WAY too much risk for me to consider this profitable!

-PatP

This would be time and materials, not flat rate. The client already buys custom software from me that interacts with the package.
My real predicament is whether there is enough work to justify the up front learning. And whether it is easier to learn another environment, or find another client that fits what I currently know.

If it was time and materials, and it was clear up front in the statement of work or the contract that you would simply implement changes as directed by the client (not professing any tax or accounting knowledge), then the risk would be acceptable to me. The part that only you can decide is whether it is time to find a new client or if the work needed is justifiable. Keep in mind that you might be able to convince the client to either pay for training or allow hours for you to train yourself, either of those could make the situation a lot more palatable.

As I remember it, you're in a pretty remote location. Finding new clients in your neighborhood might be challenging. A good client is worth a fair bit of sacrifice any day any where, so in your case that might be enough to "tip the scale" toward staying where you are.

-PatP

Decided not to do it. After talking to the client, there isn't enough work to justify the investment.